[ntp:questions] Re: Server stopped syncing to PPS?

David L. Mills mills at udel.edu
Fri Feb 3 15:07:43 UTC 2006


It is not only a problem only with non-development versions, it is a 
problem with development versions as well. There is a nonsurmountable 
issue about what it means to be correct and not. The code calculates an 
offset interval over which a source can be considered correct. It then 
finds the best intersection of these intervals according to theoretical 
principles. With so many sources and expected wiggles due to many causes 
sometimes the intersection does not contain the PPS signal. You either 
need to constrain the sources to those with low-wiggle deviation or 
increase the intersection interval padding with the tinker mindist command.


Terje Mathisen wrote:

> Peter Eriksson wrote:
>> How can one find out *why* a server suddenly decieded that the
>> PPS signal should be ignored (even though it looks just fine)
>> after having been running fine for a day or so? (It choose the PPS
>> source after a little while just as it should behave).
> This was a known problem with older (non-development) ntpd versions, 
> where a sufficient number of external sources could relatively easily 
> overwhelm/outvote a perfect PPS source.
> Terje

More information about the questions mailing list